Adaptive Educational Hypermedia Systems in Technology Enhanced Learning: A Literature Review

Catherine Mulwa, Seamus Lawless, Mary Sharp, Inmaculada Arnedillo-Sanchez, Vincent Wade

Knowledge and Data Engineering Group

School of Computer Science and Statistics Trinity College, Dublin

Telephone: 00 353 1 8961335

mulwac@scss.tcd.ie, seamus.lawless@scss.tcd.ie, mary.sharp@scss.tcd.ie, macu.arnedillo@scss.tcd.ie, vincent.wade@scss.tcd.ie,

ABSTRACT This literature review focuses on the educational benefits afforded to learners by Technology Enhanced Learning Environments (TELE) which adapt and personalize the learning experience. More specifically it focuses on Adaptive Educational Hypermedia Systems (AEHS) that incorporate Learning Styles. Adaptive approaches to learning offer alternatives to the traditional “One-size-fits-all” approach and have driven the development of dynamic educational environments. The use of such environments can deliver educational benefits as educational offerings are personalised based upon various characteristics of individual learners. Categories and Subject Descriptors D.3.2 [Computers and Education]: Computer and Information Science Education – Information Systems Education. D.3.3 [Programming Languages]: H.5.4 [Information Interfaces and Presentation]: Hypertext / Hypermedia; D.1.2 [Information Systems]: User Systems (Human Factors);

General Terms Human Factors

Keywords Technology Enhanced Learning, Adaptive Educational Hypermedia Systems, Learning Styles

1. INTRODUCTION Traditional Technology-Enhanced Learning (TEL) systems offer very few strategies for the personalisation of educational offerings. This limits the scope for providing tailored, effective TEL experiences to learners. However, adaptive educational

hypermedia systems (AEHS) have been developed to address learner dissatisfaction by attempting to personalise the learning experience. Recent research in TEL has focused on the provision of adaptive educational experiences that are tailored to the particular needs of a learner. This adaptivity can be based upon various characteristics of the learner, including knowledge level, goals or motivation. The purpose of such adaptive educational offerings is to maximize learner satisfaction, learning speed (efficiency) and educational effectiveness [1]. This literature review and paper tackles the question of: “What are the educational benefits afforded to the learners by TELE which adapt and personalize the learning experience; in particular AEHS that incorporate Learning Styles?” For the purpose of this paper, an adaptive system refers to a system which tailors its output, using implicit inferences based on interaction with the user [2]. An adaptive hypermedia system (AHS) refers to any hypertext and hypermedia system which reflects some features of the user in a user model and applies this model to adapt various visible aspects of the system to the user [3]. In other words, an AHS should be able to satisfy three criteria: it should be a hypertext or hypermedia system; it should have a user model; and it should be able to adapt the hypermedia using this model. Many AHS exceed this basic stated criterion by adding multiple models (e.g. content model, navigation model, presentation model, device model etc). AEHS have been found to be useful in engaging the learner more in the educational experience. The rest of this paper is structured as follows: Section 2 provides an overview of TEL and briefly introduces the mapping of educational theory to the design of learning environments; Section 3 introduces current AEHS; Section 4 introduces learning styles (LS) and provides a chronological taxonomy of recent research into learning style; Section 5 provides examples of AEHS which incorporate learning styles and also presents a list of potential benefits, limitations and pitfalls of AEHS. Finally Section 6 concludes the paper and recommends future work. 2. TECHNOLOGY-ENHANCED LEARNING The process of learning in formal education no longer takes place solely in traditional, educator-centric settings. Interactive, learner- centric experiences are being used to support learner

 

Permission to make digital or hard copies of all or part of this work forpersonal or classroom use is granted without fee provided that copies arenot made or distributed for profit or commercial advantage and thatcopies bear this notice and the full citation on the first page. To copyotherwise, or republish, to post on servers or to redistribute to lists,requires prior specific permission and/or a fee. SIGITE’10, October 7-9, 2010, Midland, Michigan, USA. Copyright 2010 ACM 978-1-4503-0343-9/10/10…$10.00.

73

collaboration, knowledge acquisition and reflection. Learner enquiry, activity and engagement are key requirements in such experiences and TEL applications are being designed and utilised to meet these requirements [4]. TEL practices cater to students and teachers who use many different learning tools and environments and have experience of interaction derived with open, ubiquitous, and socially-oriented services. TEL is becoming well established in higher education institutes, most notably in blended or hybrid learning scenarios, which blend TEL and traditional approaches to teaching and learning [5]. TEL is expected to make a radical difference to education, specifically, the quality and effectiveness of the learning experience with one of its key contributions being ‘personalised learning’. TEL methods have been known to change the deployment of the most important resource in the education system: teachers’ and the learners’ time [6]. Learning content should be reusable and accessible to everyone (with the author and originating institutions permission); however knowledge about individual learners should certainly not be accessible to everyone. Major benefits of TEL identified in literature range from cost savings to performance and strategic benefits [7]. In most cases learners using these technologies are able to receive instant and personalised feedback, active engagement, reusable learning materials and a safe environment where one can learn from one’s mistakes and be able to access huge amounts of beneficial material on-demand. These technologies make learning more flexible in terms of time, space and place. One of the main goals of the TEL development is to enable interoperability between different systems. Technology enhanced learning systems (TELS) that deal with learning styles are a special category of adaptive educational systems (AES), that focus on students’ learning preferences as the adaptation criterion. 2.1 Mapping educational theory to the

pedagogical design of learning environment

Recent developments in technology, coupled with the growing availability of low-cost or no-cost educational materials of high-quality (e.g., open content) have made it possible to develop powerful, yet potentially widely available technology enhanced learning environments (TELE). These environments have been increasingly studied as a way to provide a rich, supportive instructional system to students [8]. The TELE environments deliver instructional content and provide an array of scaffolding designed to support student learning. To ensure that the activities offered by learning environments can achieve the desired learning outcomes, there must be some mapping between the learning theories and the pedagogical design of learning environments. Each environment must provide specific functionality which reflects the approach to knowledge acquisition of the learning theory, or theories, which are being used. Mayes et al (2004) and Lawless (2009) provide categorization of learning at a theoretical level by dividing the1 process of learning into three broad and overlapping perspectives: i) the associationist / empiricist perspective, which defines learning as an activity, ii) the cognitive perspective, which defines learning as achieving understanding and iii) the situative perspective, which defines learning as a social practice. Each environment must provide specific functionality which reflects the approach to knowledge acquisition of the learning theory, or theories, which are being

used [4, 9]. Figure 11 below presents a display of a sample mapping of some approaches to TEL onto the pedagogical strands [9].

Figure 11 TEL models by pedagogical category

3. Using adaptive hypermedia systems for education

A hypermedia application offers its learners much freedom to navigate through a large hyperspace. Adaptive hypermedia (AH) offers its learners personalized content, presentation, and navigation support. Knutove et al (2009) provide a comprehensive overview of AH methods and techniques since their introduction 12 years ago. The researchers presented a survey of adaptive hypermedia (AH) architecture, defined a new taxonomy of adaptation techniques and also introduced a set of requirements and a modular structure that can be used to update the first generic AH model adaptive hypermedia application model (AHAM) that was introduced 10 years ago [10]. Adaptive hypermedia systems are being increasingly employed for educational purposes, especially with the advent of distance and distributed learning. One of the fundamental tenets of

1 This diagram was taken from http://www.scss.tcd.ie/seamus.lawless/papers/thesis.pdf

74

Figure 22 the Hierarchy of Underlying Factors of AEH

education is that students are different and hence learn in a variety of different ways [11]. In most cases some of these differences may be due to preferences for certain ways of working. The material used for pedagogical purposes should be adaptive (or adaptable), in order to cater for these differences. AH is a technique used to provide a personalised learning experience that draws on computer-driven intelligent tutoring systems (ITSs) and student-driven virtual learning environments (VLES).

3.1 Adaptive hypermedia in educational systems

AEHS systems offer an alternative to the non-individualized instruction approach, by providing various services adapted to the learner profile. These systems are based upon user models which characterise each individual and can use these models to offer learners educational experiences which fit their needs. To achieve this, AEHS are comprised of several sub-components which have their own distinct behaviours and properties. Figure 22 conceptually depicts the hierarchy of this scenario. For example the AEHS as the intra-artifact system shown in Figure 2 can be decomposed by considering the influence of sub-components on the performance 2 This diagram is taken from the educational wide society

(http://nccur.lib.nccu.edu.tw/bitstream/140.119/14993/1/52.pdf)

of ones at higher levels. The dashed arrows are edges representing uncertain influence. In this model the uncontrolled factors are identified and linked to other sub-components in the system with dashed arrows. For example sub-components inside the intra-artifact system can play the role of uncontrolled factors in the evaluation. On the other hand, the hierarchy of the evaluation task is shown by the evaluation-wide system. A typical architecture of the state of the art of AEHS is fully decoupled and consists of five complementary models: i) The domain model which specifies what is to be adapted ii) The user and context models which indicate what parameters the content can be adapted and iii) the instructional and adaptation models which express the pedagogical approach the learning process should be based on, as well as the forms of adaptation to be performed [12]. In a review conducted by Karampiperis et al (2005), the authors identified the current state of the art adaptive hypermedia systems as AHA! [13], OntoAIMS [14], the Personal Reader [15], WINDS [16], ACCT [17]. These systems are based on the AHAM. This model builds upon the Dexter model, that is, a common model for hypertext-based systems that was designed for general purpose adaptive web application. The model consists of two main layers: i) the run-time layer which contains the adaptation engine that performs the actual adaptation and ii) the storage layer, which stores information about the media space, the domain model, the user model and the adaptation model [12]. The main components of the AHAM model and their structural interconnections are illustrated by Figure 3 below. The dashed lines represent a logical connection between the linked models. The student model maintains an accurate representation of a student’s current state of knowledge, which allows the system to perform some adaptation based on the knowledge acquired during

75

the learning process [18]. It includes information referring to the specific knowledge that the system judges that the user possesses on the domain, known as the domain dependent data (DDD). In the past, AHS systems have attempted to customize courses to a learner’s prior knowledge, goals and personal preferences without taking into account any form of pedagogy. As a result, such systems neglect the entire body of research that exists in the educational field and fail to take advantage of the benefits that the application of pedagogy has for the learning experience [19]. The researchers’ emphasize the significance of iClass (iClass, 2004) which is an open learning system which utilizes pedagogical

strategies to adapt to learners’ needs, both intelligently and cognitively. Jovanovica et al. (2009) discuss the three generations of AEHSs. The first generation comprised stand-alone systems with adaptation rules and content entwined in a single model. They used this model together with the user model to offer personalised content (e.g. AHA! and ELM-ART). However, as adaptation rules and content were intertwined, there was little scope for content or model reuse or the use of externally developed content in the generation of learning offerings.

Figure 3 Generalized architecture of adaptive educational system

The second generation attempted to overcome some of the problems encountered by the first generation by pursuing a multi-model approach. This approach assumed decoupling of content and the adaptation rules of the system [20]. The third generation is moving towards a service-oriented architecture and the complete decoupling of different kinds of knowledge [21]. Peter Brusilovsky (2004) provided a subjective overview of research in adaptive educational hypermedia and summarized the current state of the art of the three generations [22]. He admitted that there were problems encountered while using the AEHS and

accepted that several research teams had recognized the problems of static hypertext in different application area and had begun to explore various ways to adapt the behavior of hypertext and hypermedia systems to users. For example he accepts there are problems which are related to hypermedia such as, navigation in hypermedia, inefficient navigation or the problem of being lost in hyperspace; that had been discovered when the field of hypertext reached relative maturity at the end of the 1980’s [23].

Educational Content Presenter

Adaptation Rule Parser

Behavior Tracker

   Educational Resource

Educational Resource Description Model

Learning Goals Hierarchy

Domain Concept Ontology

Media SpaceDomain Model

Adaptation Model

Content Selection

Rules

Content Selection

Rules

Learner Cognitive Characteristics and Preferences

Learner Knowledge Space

Runtime Layer Storage Layer

Figure 3: Generalized Architecture of Adaptive Educational Systems

76

3.2 Evaluations of AEHS Evaluation of any system is important and should ensure the correct methods are used. In particular it is very significant to evaluate the entire AEHS both from a technological perspective and from a user-centered perspective. This is emphasized more from our earlier research on system evaluation [24]. The evaluation of learner and tutor feedback is essential in the production of high quality personalised TEL services. There are a few evaluations available in the AH domain relative to the amount of research interest this domain is attracting. Majority of the research in this domain focuses on the technological design and performance of systems without justifying the designs through the lessons learned from evaluations [25]. In order to provide the best support for learners, a user-centered evaluation approach for enhancing and validating the student model of AEHS has been proposed, that combines adaptive hypermedia (AH) and information retrieval techniques [26]. User-centered evaluation (UCE) can serve three goals: verifying the quality of an AEHS, detecting problems in the system functionality or interface, and supporting adaptivity decisions. These functions make UCE a valuable tool for developers of all kinds of systems, because they can justify their efforts, improve upon a system or help developers to decide which version of a system to release. The benefits of the user-centered approach are savings in terms of time and cost, ensuring the completeness of system functionality, minimizing required repair efforts, and improving user satisfaction. This may lead to higher adoption of the AEHS, ease of use and a more enjoyable student experience. Student model performance is usually measured in terms of actual and expected accuracies, where actual accuracy is a model’s probability of a correct response averaged across all users. For example, Corbett and Anderson (2008) used correlation, mean error and mean absolute error to quantify model validity [27]. 4. LEARNING STYLES To date no single definition of the term learning style (LS) has been identified, a widely accepted definition is given by Keefe (1979) who defines learning style as “the composite of characteristic cognitive, affective, and physiological factors that serve as relatively stable indicators of how a learner perceives, interacts with, and responds to the learning environment” [28]. Learning styles are preferences for information types (concrete vs. abstract), presentation styles (visual vs. verbal, written) and learning actions such as hands-on versus planning and reflecting about a concept [29]. Some researchers view learning styles as individual, stable and predictable [30] others view learning styles as the parts of personality that change over time [31-32] or that unconsciously adapt to match learning contexts [33]. Both views seem acceptable, as Cassidy stated; learning styles are regarded as comprising of three fundamental learning components: i) information processing, ii) instructional preference, and iii) learning. However other researchers disagree; Elizabeth Brown (2007) is quite negative about learning styles; a quote from her thesis “Overall no statistically significant benefits were found and these findings now shed doubt as to whether learning styles are indeed an effective mechanism for personalised learning” [11]. There is no doubt that adaptivity is a good thing, it comes down to i) what properties of a learner do you adapt on and ii) are learning styles a good and effective means of teaching / learning or iii)

does adapting to learning style benefit the learner or iv) do we know how to successfully adapt to learning styles. The authors acknowledge personal learning style as being significant since it is one of the factors that influence learning. Other factors that influence learning are presented in Figure 4 below.

Figure 4 Factors that influence learning

      Learning

In a review conducted by Penger & Tekav (2009), the researchers identified the most influential research studies and research construct within learning styles. The researchers categorized 32 studies from 2000 to 2008 as shown in Table 14. However of the 32 studies only 2 were on adaptive learning systems. In their review the researchers fail to identify which of these two adaptive systems were AEHS. This supports our point that there is limited research in area of AEHS and more research and publications is required. 4.1 Measurements for Learning Styles Hsieh et al (2007) acknowledge that measurements can be used as a decision making tool. The researchers also admit that learning styles measurements are like foundations for predicting how the student will perform. The students’ different learning styles should be measured along with different information processing stages (e.g. sensory-term memory, short-term memory and long-term memory). For instance during the sensory memory stage which is affiliated with the transduction of energy (i.e., change from one energy to another), in the process of transduction a memory is created. It is critical that the learner initially process the information at this beginning stage in whole learning process (from sensory register, working memory to long -term memory).3 The students’ different learning styles should be measured along with different information processing stages (e.g. sensory-term memory, short-term memory and long-term memory).

* Factors in TELE that influence learning

77

Table 14 Chronological taxonomy of recent research into learning styles and (organizational) learning

Source: authors; adapted from the research papers and publications indicated in the figure, 2008 * Pure adaptive (educational & Learning system4

4 Table taken from http://www.cluteinstitue-onlinejournals.com/PDFs/1611.pdf

78

Figure 55 Possible Locations of Learning, Learning Measurements & Information processing stages

This process is demonstrated clearly in Figure 55 above. Actually in the past thirty or more years, the outcomes of the learning style measurement have encouraged teachers to understand students’ learning style so that they know how to enhance the students to conclude that they need to find the relationship between learning style and varied instructions, so that they can develop customized materials for students who have different learning styles [32, 34-35]. Figure 55 is the assumption where possible location of learning styles and measurements might be located. 5. AEHS INCORPORATING LS Learning styles with other means of adaptivity (e.g. user goals, prior knowledge) provide some improvements in learner satisfaction and knowledge gain. However there are very few studies in adaptive e-Learning which limit the adaptivity to just learning style adaptivity. From the results of around 2009, both of the researchers (Brown and Kelly) came to similar conclusions which seemed to indicate that the aspects of specific learning style adaptation have no significant impact on the learners performance

[36-37]. One reason for this might be that adapting to learning styles means adapting to only one feature of the learner and that the real power and benefit of AEHS systems is the dynamic personalisation of the content and the navigation based on multiple contextual influences (e.g. prior knowledge). Most of the AEHS which incorporate learning styles are based on the notion that matching the learning strategies with learning styles improves learner’s performance. Examples of these systems include AEHS which address learners’ diverse needs to assimilate and comprehend information or content [18, 38-39]. Table 2 provides a summary of examples of AEHS systems, their approach to determining learning style, the type of learning styles incorporated in these systems and a brief overview of how the AEHS system achieves its learning style and adaptivity5

5 Some parts of the diagram were extracted from P.H. Hsieh et al. 2007

79

Table 2Adaptive educational hypermedia systems considering learning styles

80

The process of building and updating the student model (i.e., classified into collaborative and automatic) is known as student modeling. The student model aid and forms a very significant role in AEHS systems. In all the systems reviewed in Table 2, they all store the learning style information in student model. The model includes all relevant information that the AEHS has gathered about the learner. This data is then used as a basis for providing suitable adaptivity. The authors acknowledge that most AEHS systems research focuses on adapting to user features such as goals / tasks, knowledge, background, hyperspace experience, and interests5. Papanikolaou and Grigoriadu (2004) acknowledge that the most important thing in exploiting different learning style categorization in AEHS is their potential to support and enhance adaptation providing appropriate guidance for AEHS developers [34]. The authors admit that there is need for an investigation on how learning style categorization could assist the design of different adaptation technologies. For example in ARTHUR, iWeaver, CS388 and MANIC the adaptation is achieved by providing different media representations for each learner. On the other hand ARTHUR and iWeaver are very similar in choice of learning styles representation. While AEC-ES provides field-dependent learners with navigational support tools and guide them through the learning material via adaptive navigation support and enables learners to switch between different instructional strategies [35]. This shows that different systems can adapt based upon learning styles using techniques such as: content adaptation, navigation paths and use of multiple navigation tools.

5.1 Potential Benefits, Pitfalls and Limitations of AEHS 5.1.1 Potential benefits of AEHS Many benefits can be attached to AEHS. For example adaptive educational games (AEG) like the adaptive learning in games through non-invasion (ALIGN) [40] can encourage learner participation, improve motivation and experiences and reinforce these with a personalised learning experience. Following is a list of examples of current benefits: These systems provide specific navigation aid, Selection of content, Metadata cognition and Provide results of the learning style tests [24]. Change learning explicitly and provide scrutability. In a survey conducted by Harrigan et al. (2009), the researchers identified some of the benefits as: Reusability,

Provision of relevant learning materials which are personalised to specific learner,

Efficiency of the AEHS systems which are user specific, Student motivation, Avoidance of information overload, Automation, flexibility and Monitory and temporal and spatial relevance.

5.1.2 Limitations and Pitfalls of AEHS It is difficult to find hard evidence for the impact of the new

technologies on learning outcomes and Effective learner modelling, addressing personalisation for

disabilities considering the time and evolving user context(s), including the user control and relating them to the issues of privacy,

The complexity of environments in which learners work. Their attitudes, motivation, beliefs, knowledge and skills constitute just one important factor relevant to their learning achievements. Kay (2008) advocates that scrutability is an issue crucial not just for student models but all components of AEHSs.

Several researchers have identified some pitfalls encountered by developers of these systems [41-43]: Specification of control conditions, these occur when the

control conditions of experimental settings for evaluation are defined. In many studies the adaptive system is compared to a non-adaptive version of the system with the adaptation switched off,

Difficulty in attributing cause: is the adaptation causing the measured effect or another aspect of system functionality or design (e.g. system usability),

Statistically insignificant results, Difficulty in defining the effectiveness of adaptation, Insufficient resources and allocation of these resources, in

most cases the resources are underestimated and Too much emphasis on summative rather than formative

evaluation.

Meccaway et al (2008) emphasized that the level adoption of AEHS into ‘real world’ teaching has being poor. The researchers agreed that one of the reasons behind this is due to their architectural design failing to answer the overall needs of Web-enhanced learning [44]. Majority of these systems are currently prototypic and experimental systems with basic graphical user interface (GUI). They are not designed for the modern e-Learning context which prompts services and reusability of learning context [45]

81

Table 3 Adaptive and Adaptive hypermedia variables identified in the studies

The authors have identified existing adaptive variables variable (also known as concepts) that can prompt adaptivity, in the literature from 1996 to 2008. These variables make AEHS a variable tool for learners in TELE. By adaptive variables we refer to the features of the user that are used as a source of the

adaptation (i.e. to what features of the user the system can adapt its behaviour). Table 3 above presents a summary of these variables, the researchers who identified them and the year [2-3].

82

6. CONCLUSIONS AND FUTURE WORK Nowadays, the development of technology enriches the learning environment where the teachers can continuously monitor the appropriateness of their instructional delivery methods [42]. It remains to be seen if educators are willing to invest the significant time and effort required to initially integrate an AEHS systems into their teaching plans and to ensure that the learning offerings provided meet the curriculum. The teachers can review information processing theories and cognitive theories, and then examine the structure of learning style measurements. The paper demonstrated how the use of TEL, particularly AEHS, can benefit learners and tackled our research question “What are the educational benefits afforded to learners by technology- enhanced learning environments (TELE) which adapt and personalize the learning experience. More specifically it focuses on adaptive educational hypermedia Systems (AEHS) that incorporate learning styles?” It emphasized the importance of incorporating learning styles into AEHS in order to provide a more personalised and effective learning experience although some researchers may disagree i.e.,[46] [11]. Adaptivity can only be provided in an efficient and effective manner if the needs of the students are known. The landscape of AEHS is rather rich in ideas and interesting solutions. The authors came to conclusion that; learning styles with other means of adaptivity (e.g. user goals, prior knowledge) provide some improvements in learner satisfaction and knowledge gain. However there are very few studies in adaptive e-Learning which limit the adaptivity to just learning style adaptivity. Some researchers seemed to indicate the aspects of specific learner’s performance; adaptation has no significant impact on learner’s performance. One of the reasons for this might be that adapting to learning styles means adapting to only one feature of the learner and that the real power and benefit of AEHS systems is the dynamic personalisation of the content and the navigation based on multiple contextual influences (e.g., prior knowledge). AEHS incorporating learning styles are very significant to learners. Further work is required to address the limitations of such systems and to: Investigate the weaknesses of the AEHS which incorporate

Learning styles; Investigate how such systems have been evaluated and

identify the most effective approaches; Compile a review of all existing AEHS; Examine the extent to which the decision to allow different

levels of control over the student model is determined by the domain.

7. ACKNOWLEDGEMENTS

This research is based upon works supported by Science Foundation Ireland (Grant Number: 07/CE/I1142) as part of the Centre for Next Generation Localization (www.cngl.ie). The authors are grateful for the suggestions of the reviewers for this paper.

8. REFERENCES [1] E. Popescu, et al., "Towards a unified learning style model in

adaptive educational systems," Proc. ICALT 2007, pp. 804-808, 2007.

[2] L. Van Velsen, et al., "User-centered evaluation of adaptive and adaptable systems: a literature review," The knowledge engineering review, vol. 23, pp. 261-281, 2008.

[3] P. Brusilovsky, "Methods and techniques of adaptive hypermedia," User Modeling and User-Adapted Interaction, vol. 6, pp. 87-129, 1996.

[4] S. Lawless, "Leveraging Content from Open Corpus Sources for Technology Enhanced Learning," PhD, School of Computer Science & Statistics, Trinity College Dublin, Dublin, 2009.

[5] U. Schroeder. (2009). Web-Based Learning–Yes We Can! 5686/2009.

[6] D. Laurillard, "Modelling benefits-oriented costs for technology enhanced learning," Higher Education, vol. 54, pp. 21-39, 2007.

[7] C. Rainsford and E. Murphy, "Technology-enhanced learning," Journal of European Industrial Training, vol. 29, 2005.

[8] A. J. Girasoli and R. D. Hannafin, "Using asynchronous AV communication tools to increase academic self-efficacy," Computers and Education, vol. 51, pp. 1676-1682, 2008.

[9] T. Mayes and S. De Freitas, "Review of e-learning theories, frameworks and models," JISC E-Learning Models Study Report, 2004.

[10] E. Knutov, et al., "AH 12 years later: a comprehensive survey of adaptive hypermedia methods and techniques," New Review of Hypermedia and Multimedia, vol. 15, pp. 5-38, 2009.

[11] E. Brown, "The use of learning styles in adaptive hypermedia," PhD School of Computer Science & IT, University of Nottingham, 2007.

[12] P. Karampiperis and D. Sampson, "Adaptive learning resources sequencing in educational hypermedia systems," Educational Technology & Society, vol. 8, pp. 128-147, 2005.

[13] P. De Bra, et al., "Aha! version 2.0, more adaptation flexibility for authors," 2002, pp. 240-246.

[14] L. Aroyo, et al., "OntoAIMS: ontological approach to courseware authoring," 2003, pp. 2-5.

[15] P. Dolog, et al., "The personal reader: Personalizing and enriching learning resources using semantic web technologies," in Third International Adaptive Hypermedia and Adaptive Web-based Systems Conference, Eindhoven, The Netherlands, 2004, pp. 85-94.

[16] M. Kravcik and M. Specht, "Flexible navigation support in the winds learning environment for architecture and design," 2004, pp. 156-165.

[17] D. Dagger, et al., "Personalisation for all: Making adaptive course composition easy," Educational Technology & Society, vol. 8, pp. 9-25, 2005.

83

[18] S. Graf, "Adaptivity in Learning Management Systems Focussing on Learning Styles," Vienna University of Technology, 2007.

[19] I. O'Keeffe, et al., "Just-in-time Generation of Pedagogically Sound, Context Sensitive Personalized Learning Experiences", International Journal on E-Learning (IJeL), Special Issue: Learning Objects in Context," International Journal on E-Learning (IJeL), vol. 5, pp. 113-127, 2006.

[20] O. Conlan, "The multi-model, metadata driven approach to personalised eLearning services," PhD, Department of Computer Science, Trinity College, Dublin, 2005.

[21] J. Jovanovica, et al., "The Social Semantic Web in Intelligent Learning Environments: state of the art and future challenges," Interactive Learning Environments, vol. 17, pp. 273-309, 2009.

[22] P. Brusilovsky, "Adaptive navigation support: From adaptive hypermedia to the adaptive web and beyond," PsychNology Journal, vol. 2, pp. 7-23, 2004.

[23] P. Brusilovsky, "Adaptive educational hypermedia: From generation to generation," 2004, pp. 19-33.

[24] C. Mulwa, et al., "OSSES: An Online System for Studies on Evaluation of Systems," presented at the ED-MEDIA 2010-World Conference on Educational Multimedia, Hypermedia & Telecommunications, Toronto, Canada, 2010.

[25] O. Conlan and V. Wade, "Evaluation of APeLS - An Adaptive eLearning Service based on the Multi-model, Metadata-driven Approach," presented at the Third International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems (AH2004) Proceedings, Eindhoven, The Netherlands, 2004.

[26] Séamus Lawless, et al., "A Proposal for the Evaluation of Adaptive Personalized Information Retrieval," presented at the CIRSE 2010 Workshop on Contextual Information Access, Seeking and Retrieval Evaluation, Milton Keynes, UK, 2010.

[27] M. Yudelson, et al., "A multifactor approach to student model evaluation," User Modeling and User-Adapted Interaction, vol. 18, pp. 349-382, 2008.

[28] J. Keefe, "Learning style: An overview," Student learning styles: Diagnosing and prescribing programs, pp. 1-17, 1979.

[29] L. Obal, et al., "The influence of cognitive mapping, learning styles and performance expectations on learning effectiveness," 2007, pp. 1-20.

[30] P. Fahy and M. Ally, "Student learning style and asynchronous computer-mediated conferencing (CMC) interaction," The American Journal of Distance Education, vol. 19, pp. 5-22, 2005.

[31] D. Salter, et al., "A longitudinal study of learning style preferences on the Myers-Briggs type indicator and learning style inventory," Journal of College Student Development, vol. 47, pp. 173-184, 2006.

[32] A. Wintergerst, et al., "Conceptualizing learning style modalities for ESL/EFL students," System, vol. 31, pp. 85-106, 2003.

[33] A. Honigsfeld and R. Dunn, "High school male and female learning-style similarities and differences in diverse nations," The Journal of Educational Research, pp. 195-206, 2003.

[34] P. HSIEH, et al., "Rationales of Providing Adaptive Educational Systems in Enhancing Different Learning Style Students’ Reading Comprehension Abilities," 2007.

[35] L. Price, "Individual differences in learning: Cognitive control, cognitive style, and learning style," Educational Psychology, vol. 24, pp. 681-698, 2004.

[36] E. Brown, et al., "Patterns in authoring of adaptive educational hypermedia: a taxonomy of learning styles," JOURNAL OF EDUCATIONAL TECHNOLOGYAND SOCIETY, vol. 8, p. 77, 2005.

[37] D. Kelly and B. Tangney, "Learners ‘Responses to Multiple Intelligence Differentiated Instructional Material in an ITS," Artificial intelligence in education: shaping the future of learning through intelligent technologies, p. 446, 2003.

[38] T. Wang, et al., "Using a style-based ant colony system for adaptive learning," Expert Systems with Applications, vol. 34, pp. 2449-2464, 2008.

[39] E. Sangineto, et al., "Adaptive course generation through learning styles representation," Universal Access in the Information Society, vol. 7, pp. 1-23, 2008.

[40] N. Peirce, et al., "Adaptive educational games: Providing non-invasive personalised learning experiences," in Proceedings of the Second IEEE International Conference on Digital Games and Intelligent Toys Enhaced Learning, Banff, Canada, 2008, pp. 28-35.

[41] S. Lawless, et al., "A Proposal for the Evaluation of Adaptive Personalised Information Retrieval," in Proceedings of the 2nd International Workshop on Contextual Information Access, Seeking and Retrieval Evaluation

Milton Keynes, UK, 2010. [42] N. Tintarev and J. Masthoff, "Evaluating Recommender

Explanations: Problems Experienced and Lessons Learned for the," UMAP 2009, p. 54, 2009.

[43] C. Gena and S. Weibelzahl, "Usability engineering for the adaptive web," The Adaptive Web, pp. 720-762, 2007.

[44] M. Meccawy, et al., "WHURLE 2.0: Adaptive Learning Meets Web 2.0," Times of Convergence. Technologies Across Learning Contexts, pp. 274-279, 2008.

[45] P. Brusilovsky, et al., "From Learning Objects to Adaptive Content Services for E-Learning," Architecture Solutions for E-Learning Systems. IGI Global, pp. 243–261, 2007.

[46] E. Brown, "The use of learning styles in adaptive hypermedia," PhD School of Computer Science & IT, University of Nottingham, 2007.

84